CMV infection of hematopoietic cell transplant (HCT) recipients is a continuing problem that impacts the outcome of this very successful therapy. Anti-viral treatment with ganciclovir/foscarnet is the main treatment strategy to prevent CMV disease post-transplant (Tx). Despite significant advances in formulation and delivery of anti-virals, their use complicates and extends the post-Tx recovery period and risk for CMV disease. Considering these caveats, we are pursuing a novel therapeutic strategy that focuses on priming or enhancing CMV-specific T cell immunity in healthy volunteers and HCT donors. In developing this vaccine strategy, it is assumed that the target population will be immunocompetent, and vaccination judged successful if immunity to CMV in response to the vaccine is equivalent in individuals with self-limited CMV infection, which at first approximation represents a protective response. Towards this end in Specific Aim (SA1) attenuated poxviruses (MVA) will be constructed that express four CMV antigens (CMV-MVA) including UL32, UL44, UL83, and UL123-exon4 for stimulation of cellular immunity to CMV. MVA has several advantages including avirulence, low inflammatory response and pathogenicity in humans, and inclusion of multiple CMV antigens will ensure broad vaccine coverage for the USA. Lack of viral assembly in mammals, together with studies showing its safety in heavily immunosuppressed macaques, rodents, and in HIV-AIDS patients is evidence of its candidacy for use in HCT recipients. CMV-MVA constructed in SA1 will be qualified as immunologically functional, and subsequently transferred to a certified cGMP manufacturer for clinical production. To ensure safety of the rMVA for Tx recipients, a safety study in SA2 will be conducted in healthy CMV positive and negative volunteers. The immunization regimen will model the time frame required for providing two doses of vaccine to Tx donors without delaying Tx. A Phase II trial using a randomized balanced cohort of 140 vaccinees and unvaccinated controls is proposed in SA3 to establish whether immunization of a donor with CMV-MVA prior to Tx will result in modification of the course and magnitude of CMV-viremia in a Tx recipient. This trial will be sufficiently powered that a 62% reduction in viremia will be statistically meaningful as a measure of the success of the vaccine. Successful limitation of acute CMV infection by this approach lays the foundation for addressing the important problem of late CMV disease by immunizing Tx recipients with the same or an improved version of CMV-MVA.